Locating a user based on aggregated tweet content associated with a location

Abstract

A user submitting a query from a computer at an unknown location is located using a language model. The language model is derived from an aggregation of tweets that were sent from known locations.

What is claimed is: 
     
       1. A computer-implemented method for locating a user at an unknown location based on a language model derived from aggregated tweets sent from known locations, comprising:
 aggregating a plurality of tweets from a plurality of users at a plurality of known locations, each tweet comprising microblog content and a geotag with location coordinates of the tweet; 
 grouping microblog content of tweets sent from a common location; 
 receiving query content from a user at an unknown location; 
 estimating the unknown location from the plurality of known locations by comparing at least a portion of the query content to the microblog content; and 
 outputting the estimated location. 
 
     
     
       2. The method of  claim 1 , further comprising:
 building a histogram of word or phrase frequency for each common location. 
 
     
     
       3. The method of  claim 1 , wherein the receiving query content step occurs subsequent to the grouping microblog content step. 
     
     
       4. The method of  claim 1 , wherein the common location comprises at least one of a country, a state, city, a zip code, a neighborhood, a point of interest, a province, a grid location, or an arbitrary boundary. 
     
     
       5. The method of  claim 1 , wherein the location coordinates for a geotag comprises a longitude and latitude. 
     
     
       6. The method of  claim 1 , further comprising:
 receiving an IP address associated with the unknown location; and 
 retrieving IP-based location information, 
 wherein estimating the unknown location use the IP-based location as an input. 
 
     
     
       7. The method of  claim 1 , wherein the comparing the query content to the microblog content comprises:
 generating a list of common locations correlated to the query content; and 
 ranking the list of common locations based on a probability score that tweet text of a common location matches the tweet text of an unknown location. 
 
     
     
       8. The method of  claim 7 , wherein the predicted certain common location is selected as a highest probability score. 
     
     
       9. The method of  claim 1 , further comprising:
 providing location-based information associated with the certain common location. 
 
     
     
       10. A computer program product stored on a non-transitory computer-readable medium that when executed by a processor, performs a method for locating a user based on tweets associated with a location, the method comprising:
 aggregating a plurality of tweets from a plurality of users at a plurality of known locations, each tweet comprising microblog content and a geotag with location coordinates of the tweet; 
 grouping microblog content of tweets sent from a common location; 
 receiving query content from a user at an unknown location; 
 estimating the unknown location from the plurality of known locations by comparing at least a portion of the query content to the microblog content; and 
 outputting the estimated location. 
 
     
     
       11. The product of  claim 10 , further comprising:
 building a histogram of word or phrase frequency for each common location. 
 
     
     
       12. The product of  claim 11 , wherein the predicted certain common location is selected as a highest probability score from a plurality of potential common locations. 
     
     
       13. The product of  claim 10 , wherein the receiving query content step occurs subsequent to the grouping microblog content step. 
     
     
       14. The product of  claim 10 , wherein the common location comprises at least one of a country, a state, city, a zip code, a neighborhood, a point of interest, a province, a grid location, or an arbitrary boundary. 
     
     
       15. The product of  claim 10 , wherein the location coordinates for a geotag comprises a longitude and latitude. 
     
     
       16. The product of  claim 10 , further comprising:
 receiving an IP address associated with the unknown location; and 
 retrieving IP-based location information, 
 wherein estimating the unknown location use the IP-based location as an input. 
 
     
     
       17. The product of  claim 10 , wherein the comparing the query content to the microblog content comprises:
 generating a list of common locations correlated to the query content; and 
 ranking the list of common locations based on a probability score that a common location matches the unknown location. 
 
     
     
       18. A system for locating a user based on tweet content associated with a location, comprising:
 a tweet database; 
 a tweet training module, coupled in communication with the tweet database, the aggregate a plurality of tweets from a plurality of users at a plurality of known locations, each tweet comprising microblog content and a geotag with location coordinates of the tweet, the tweet training module grouping microblog content of tweets sent from a common location and storing in the tweet database; 
 a tweet query module, coupled in communication with the tweet database, the tweet query module to receive query content from a user at an unknown location, the tweet query estimating the unknown location by comparing at least a portion of the query content to the microblog content, and outputting the estimated location from the tweet database. 
 
     
     
       19. The system of  claim 18 , wherein the common location comprises at least one of a country, a state, a city, a zip code, a neighborhood, a point of interest, a province, a grid location, or an arbitrary boundary. 
     
     
       20. The system of  claim 18 , wherein the tweet query module receives an IP address associated with the unknown location, and retrieves IP-based location information to estimate the unknown location using the IP-based location as an input.